Client: Truck Ohkoku Co., Ltd.
Truck Inventory Search Agent
Find the right truck across hundreds — or thousands — of units, in plain language
This case study covers how Optimium Inc. built a truck-inventory search AI agent for Truck Ohkoku Co., Ltd., the company behind Truck Ohkoku (トラック王国), a Japanese used-truck marketplace. The agent matches natural-language inquiries against hundreds to thousands of trucks in live stock.
Project overview
| Item | Detail |
|---|---|
| Client | Truck Ohkoku Co., Ltd. — operator of the Truck Ohkoku marketplace |
| Timeline | ~6 months (3-month PoC + 3-month build) |
| Team | 2 Optimium members (PM-engineer / AI engineer) |
| Phases | Requirements → PoC → Production build → Operational tuning |
| Key tech | LLM agent, natural-language-to-structured-query translation, live inventory DB integration |
Problem
A used-truck distributor running hundreds to thousands of vehicles in stock was bottlenecked on the sales side. Spreadsheet filters handled simple criteria like tonnage and body type, but vague, multi-axis requests across model year, mileage, body configuration (cranes, refrigeration units, and other bed-mounted equipment), general equipment, and service history fell back on veteran staff's memory — and new hires took a long time to come up to speed.
Approach
- Structured vehicle metadata (model, year, mileage, body type, tonnage, equipment, service history) and built an agent that maps natural-language queries to structured conditions
- Connected the agent to live inventory so it can distinguish "available", "in negotiation", and "in service", and only surfaces deliverable units
- Designed a conversational UI where sales reps accept/reject candidates and narrow down iteratively
- When no unit matches exactly, the agent relaxes or reinterprets the conditions to propose close-enough vehicles the customer may still want
Outcome
- Sales reps now call back deals previously lost to "no matching truck": the agent relaxes and reinterprets the original conditions to surface vehicles the customer turns out to be interested in — reviving lost opportunities became the agent's main contribution
- Faster, more reproducible first proposals also shortened new-hire ramp-up and reduced dependence on veteran staff